Modules and import

Ways to import

import math
from math import sqrt, pi
import datetime as dt

print(math.sqrt(16), sqrt(25))
print(f"{pi:.4f}")
print(math.floor(2.7), math.ceil(2.1), math.gcd(12, 18))

4.0 5.0
3.1416
2 3 6

I prefer import modulo and use modulo.funcao, because the reader sees where each name comes from. Avoid from modulo import *: it throws dozens of names into your code and makes it impossible to know where each one came from.

The standard library

Python comes with a huge standard library, and knowing it prevents installing a package for what already exists:

import statistics
from datetime import date, timedelta

notas = [7, 8, 9, 10]
print(statistics.mean(notas), statistics.median(notas))
hoje = date(2026, 10, 6)
print(hoje.strftime("%d/%m/%Y"))
print(hoje + timedelta(days=30))
print((date(2026, 12, 25) - hoje).days)

8.5 8.5
06/10/2026
2026-11-05
80

import random

print(random.randint(1, 6) in range(1, 7))
print(random.choice(["a", "b", "c"]) in "abc")

True
True

The random is used for games and simulations. For passwords and tokens, which need to be truly unpredictable, use the module secrets.

Your own modules

Any file .py becomes an importable module. See a couple of files in the folder exemplos/modulos:

"""Funções de apoio importadas pelo programa principal."""


def saudacao(nome: str) -> str:
    return f"Olá, {nome}!"


if __name__ == "__main__":
    print(saudacao("teste do módulo"))

from util import saudacao

print(saudacao("Ana"))

python3 exemplos/modulos/principal.py

When executing a script, Python places the script's folder at the beginning of the module search list, and that’s why from util import saudacao works.

The if __name__ == "__main__" pattern

Every module has a variable __name__. When the file is executed as a program, the value is "__main__". When it is imported, the value is the module's name. This test allows the same file to be both a library and a program at the same time:

def main():
    print(f"executando com __name__ = {__name__!r}")


if __name__ == "__main__":
    main()

executando com __name__ = '__main__'

Without this protection, importing the module would execute the entire program as a side effect.

How Python finds modules

The interpreter looks in sys.path, a list of folders, in order. After being imported, the module is stored in sys.modules, and the second import simply reuses what is already there:

import sys

print(type(sys.path).__name__, "math" in sys.modules)

list True

Third-party packages come from PyPI and are installed within a virtual environment (chapter 4), with pip install, uv add or poetry add. The own packages, with __init__.py and folder structure, return in the packaging chapter.